Yea nice find and link Jerry,
Totally ties into the topic. More reading to properly get my head around, where does the bl--dy time go...
vamoose
MPPT for Stirling Engines
Re: MPPT for Stirling Engines
MPPT Oct 3, 2012
Got a simple bracket mounted to the JS002 Toy SE to hold the spindle motor from the DVD player to the flywheel. Stretched butyl inner tube sections over the JS002 flywheel and spindle motor disk to form a gravity pinch roller. After some finagling, I got it to even illuminate a LED (1.68V) and still generate 4-5mA of current (8.4mWatt). So I shot and created my first YouTube videos (see what you boys are doing to me; learning new tricks!).
[youtube]http://www.youtube.com/watch?v=iESG5BUYzgw[/youtube]
My next video was to short the motor and measure the current. Most times this would stall the JS002, but today it was running proud (I just took it apart and cleaned it), generating up to 50mA at almost 100mV as captured on the following video (5mWatt)…
[youtube]http://www.youtube.com/watch?v=kL26WyZ_kMM[/youtube]
My new tachometer should be here any day now. Then the data collection will begin. But I do get a good idea on RPM with the voltage I measure from the DVD spindle motor…
I have run this several times now and observe that the power from the setup starts off slow, goes to a peak, and then starts to die off. The videos were taken at peak performance. I noticed that this toy will essentially over heat. If I take a wash cloth with cool water, I can get back the performance if not increase performance (got over 70mA and 110mV). To this is going to be interesting for the VMPPT as this is what we want to track out and find best performance or max output...
I also noticed I could make performance worse if I cooled it too fast (ice), as well as better. Very strange little SE…
I also ordered a simple DC-DC converter that I think will help match the DVD Spindle motor output to the LED. This will be similar to a solar MPPT…
Good night!
-J
Got a simple bracket mounted to the JS002 Toy SE to hold the spindle motor from the DVD player to the flywheel. Stretched butyl inner tube sections over the JS002 flywheel and spindle motor disk to form a gravity pinch roller. After some finagling, I got it to even illuminate a LED (1.68V) and still generate 4-5mA of current (8.4mWatt). So I shot and created my first YouTube videos (see what you boys are doing to me; learning new tricks!).
[youtube]http://www.youtube.com/watch?v=iESG5BUYzgw[/youtube]
My next video was to short the motor and measure the current. Most times this would stall the JS002, but today it was running proud (I just took it apart and cleaned it), generating up to 50mA at almost 100mV as captured on the following video (5mWatt)…
[youtube]http://www.youtube.com/watch?v=kL26WyZ_kMM[/youtube]
My new tachometer should be here any day now. Then the data collection will begin. But I do get a good idea on RPM with the voltage I measure from the DVD spindle motor…
I have run this several times now and observe that the power from the setup starts off slow, goes to a peak, and then starts to die off. The videos were taken at peak performance. I noticed that this toy will essentially over heat. If I take a wash cloth with cool water, I can get back the performance if not increase performance (got over 70mA and 110mV). To this is going to be interesting for the VMPPT as this is what we want to track out and find best performance or max output...
I also noticed I could make performance worse if I cooled it too fast (ice), as well as better. Very strange little SE…
I also ordered a simple DC-DC converter that I think will help match the DVD Spindle motor output to the LED. This will be similar to a solar MPPT…
Good night!
-J
[hr] If crime doesn't pay, does that mean my job is a crime?
Re: MPPT for Stirling Engines
Nice little setup NerdyEE.
Could the curved response be a result of the thermal lag.
Once the cold sides thermal mass/sink temperature transfer, catches up with outward thermal conduction and lag, its overall temp starts to increase creating a lower Delta t. Just a thought, not even sure if my comment makes written sense....
I am working on a response that may be useful for the mppt idea. It will probably be more generic and conceptual than technically practical...
Here's a couple of possible random name ideas,
ESELO - electronic stirling engine load optimiser
DRTR - drive responsive torque regulator
feel free to suggest others.
vamoose
Could the curved response be a result of the thermal lag.
Once the cold sides thermal mass/sink temperature transfer, catches up with outward thermal conduction and lag, its overall temp starts to increase creating a lower Delta t. Just a thought, not even sure if my comment makes written sense....
I am working on a response that may be useful for the mppt idea. It will probably be more generic and conceptual than technically practical...
Here's a couple of possible random name ideas,
ESELO - electronic stirling engine load optimiser
DRTR - drive responsive torque regulator
feel free to suggest others.
vamoose
Re: MPPT for Stirling Engines
Hey J (and others), so here goes (fingers crossed)....
This post is just kind of putting some ideas out there. Fell free to respond in any way you like to it. If I have made any assumptions or points that are incorrect please raise them if you like.
For this particular post we might assume that the Stirling engine is one with a fixed value heat source and a constant pressure (be it a pressurised or non-pressurised engine), it just needs to be stabilised.
I guess the problem with finding the best output of a Stirling engine is there are so may variables and the reaction time to the thermal variations may be slow, and when you put a load on an engine the speed reduces, this results in a reduction in the amount of gas and energy processed which can also affect the delta t, and then there's thermal lag, and different forms of generators and 'bla bla bla etc'
So throwing all that in the bin for the moment her is one possible approach.
What we're trying to do is find the best load and therefore RPM of an engine in its stabilised long term operating conditions. Wether the 'regulator' overall consumes more power than it makes or vice versa doesn't matter really for the moment. its just a measuring tool. (although it would be nice to get a gain)
This process is about finding the best output of a Stirling engine and the attached generator. If you were to pair the engine with a different generator then you would have to run the process again and your result might be better or worse to a greater or lesser extent depending on how well the engine and generator are matched to each other.
So once the engine is up to operational heat and spinning freely the engine is ready for testing. I will call this stage 'open fly mode' (not like the mode, that some gents experience when exiting a bathroom)
The engine is now ready for calibration. This process will intentionally take some time due to the slow reaction of the engine as it drifts from the affects.
So we start with plotting a volt amp curve, I figure we can do this in 2 ways, one by using a rising frequency (variable frequency, with a set mark) or otherwise using pulse width modulation with progressively larger marks and lower space distances (this will have a constant/allocated frequency, - Could a lower frequency result in possibly lower reactance problems/losses?). The amplitude will be a passenger to the RPM I suspect.
We now run the engine through the cycle (this process may intentionally need to be allocated 5 minutes or more as a result of possibly slow load response time of a Stirling engine) the curve is plotted and a peak range is isolated. The regulator cuts out and allows the engine to go back into 'open fly mode' for a little bit.
The regulator then re-engages focusing in on the more refined peak output range as its new reference window for another cycle (say potentially another 5 minutes, or whatever), it is plotted, and a smaller peak output range is again identified.
Now with this new window (after another open fly mode) the regulator tries to lock back into the recorded peak output point. Once close, it slowly drifts back and forth trying to seek out the best operational load scenario . After this process, the best known RPM can be recorded for this engine/generator match, so if the engine is at operational temperature and at this RPM under a load, then you know you are close to the peak output.
If the engine stalls during the testing period you could possibly use the regulator to restart the engine via the generator. Although for different reasons, in some way similar to what Ian was indicating (sorry j, read back and realized you mentioned this as well)
These are just some ideas, and hope they make some kind of sense and are of some use...
vamoose
This post is just kind of putting some ideas out there. Fell free to respond in any way you like to it. If I have made any assumptions or points that are incorrect please raise them if you like.
For this particular post we might assume that the Stirling engine is one with a fixed value heat source and a constant pressure (be it a pressurised or non-pressurised engine), it just needs to be stabilised.
I guess the problem with finding the best output of a Stirling engine is there are so may variables and the reaction time to the thermal variations may be slow, and when you put a load on an engine the speed reduces, this results in a reduction in the amount of gas and energy processed which can also affect the delta t, and then there's thermal lag, and different forms of generators and 'bla bla bla etc'
So throwing all that in the bin for the moment her is one possible approach.
What we're trying to do is find the best load and therefore RPM of an engine in its stabilised long term operating conditions. Wether the 'regulator' overall consumes more power than it makes or vice versa doesn't matter really for the moment. its just a measuring tool. (although it would be nice to get a gain)
This process is about finding the best output of a Stirling engine and the attached generator. If you were to pair the engine with a different generator then you would have to run the process again and your result might be better or worse to a greater or lesser extent depending on how well the engine and generator are matched to each other.
So once the engine is up to operational heat and spinning freely the engine is ready for testing. I will call this stage 'open fly mode' (not like the mode, that some gents experience when exiting a bathroom)
The engine is now ready for calibration. This process will intentionally take some time due to the slow reaction of the engine as it drifts from the affects.
So we start with plotting a volt amp curve, I figure we can do this in 2 ways, one by using a rising frequency (variable frequency, with a set mark) or otherwise using pulse width modulation with progressively larger marks and lower space distances (this will have a constant/allocated frequency, - Could a lower frequency result in possibly lower reactance problems/losses?). The amplitude will be a passenger to the RPM I suspect.
We now run the engine through the cycle (this process may intentionally need to be allocated 5 minutes or more as a result of possibly slow load response time of a Stirling engine) the curve is plotted and a peak range is isolated. The regulator cuts out and allows the engine to go back into 'open fly mode' for a little bit.
The regulator then re-engages focusing in on the more refined peak output range as its new reference window for another cycle (say potentially another 5 minutes, or whatever), it is plotted, and a smaller peak output range is again identified.
Now with this new window (after another open fly mode) the regulator tries to lock back into the recorded peak output point. Once close, it slowly drifts back and forth trying to seek out the best operational load scenario . After this process, the best known RPM can be recorded for this engine/generator match, so if the engine is at operational temperature and at this RPM under a load, then you know you are close to the peak output.
If the engine stalls during the testing period you could possibly use the regulator to restart the engine via the generator. Although for different reasons, in some way similar to what Ian was indicating (sorry j, read back and realized you mentioned this as well)
The output from the regulator may run through a dc to dc converter and then to a battery (possibly with some kind of resistive load dump to stop the battery from reaching a full state of charge, float voltage.)Ian S C wrote:The old system for petrol(gas), chargers/ power suplies was the battery voltage dropped in use, when it got to a pre determand level, a relay dropped out, and the battery switched to the generator causing it to motor over, starting the petrol motor, when the revs came up the generator started charging, and continued until the battery was up to charge, and the motor was shut down, all automatic, and no electronics, just relays. Could be done electronicly, and give a more scientific charge rate, could even be designed for Lithium batteries. Ian S C
These are just some ideas, and hope they make some kind of sense and are of some use...
vamoose
Re: MPPT for Stirling Engines
Hi Vamoose,
What you proposed is essentially what I have been doing this past week, measuring what my small little JS002 toy SE and the DVD spindle motor can produce in speed and power. I did approximately 20 runs and plotted 20 different RPM/Power curves. Similar trends, but all different data. So your first assumption that the heating source and mechanical are all fixed and do not change, is not a valid assumption…
The concept of the “Dither & Lock” (D&L) uses the response of the SE (bell shaped RPM/Power curve) as an error signal. No need to map out the physical space as the D&L will locate the max point, even if it changes due to pressure, temperature, or phase of the moon… (Even a different mechanical configuration/generator!)
The output of the D&L is a full wave rectified version of the dither frequency (see my Excel simulation). The polarity of this full wave signal will change once past the peak operating point (peak of bell curve). At the peak, the error is zero. Thus taking this error signal into an integrator (PID or the like) can drive the applied loading to the SE to maintain optimum performance. You are correct that this dither frequency needs to be less than the natural response of the SE (need the SE to respond to the change in command for power). From what I see from my toy, that frequency is sub hertz (about 0.1HZ). This may be my difficult challenge as this is a very slow frequency and thus the overall response may be even slower…
FYI- my comment about using ice to kill the toy SE motor, I figured out what was happening. The compression cylinder has an inner sleeve of cast iron (?) and so is the piston. Suspect the people that made this toy read Ian’s numerous posts on friction and the use of cast iron. What happens when I hit the toy with ice is that the compression cylinder seizes solid. But if I cool it slowly (water drips or slow ice ingress) the motor will run very fast. At the time I didn’t have my tachometer so was unable to measure how fast I was able to get the toy. I felt a water drip cooler was not a good long term solution for these tests…
So what have I done lately? I tried to improve the thermal performance of the toy SE. The toy would only allow me about 3 minutes of operation at its peak (> 1000RPM) before it would die off (< 1000RPM). Taking inspiration from Ian’s 2.5cc SE w/ generator, I mounted extra heat sinks and was then able to operate the toy for multiple alcohol bottle burns without a noticeable change in power output. Though honestly, I would rather use Ian’s 2.5cc w/ generator for these experiments as I think my toy is about to die… (have to fix something approx every 5-7 burns and thus obtain different performance)
[album]85[/album]
I think this and my alcohol burner variability is why I can’t obtain good repeatable RPM/Power curves, but I’m going to try it again once I build a true PWM load to measure the full RPM/Power curve (can only measure to the peak now). Somewhere on the forum I read to use graphite as a lubricant. If doing so, I found one has to apply the stuff like eye-shadow (very gently with Q-tip swab and not puff-puff and fly everywhere; a.k.a. MESS!). This does help each re-build and the max “OPEN FLY” performance I have to date is approx 1300RPM (1100 is typical)… (Please note: SE Toy flywheel has 4 holes so tachometer data needs to be corrected by dividing value on meter by 4)
I did find one thing that may be of interest to those who wish to attempt a real Solar MPPT with their SE generator setup. As mentioned before, I ordered a DC-DC converter demo board (Texas Instruments TPS61202EVM-179) that will take the 0.8 – 1.5V output from the SE and DVD Spindle motor combo, and generate a regulated 5V output. I then made a LED constant current source using a LM317 (see picture & videos) to vary the loading on the SE. What I found is that once the power demand from the load was greater than the supply from the SE (approx 12mW), the system would get crow-barred and stall the SE. I somewhat expected this behavior and thus my comments about the use of a Solar MPPT in a previous post… (I can explain what and why this is happening if people are interested in more of my babblings; constant power demand from DC-DC reduces voltage out of SE generator, thus increasing demand for current, thus further reducing voltage, thus further increasing demand for current… Falling off a cliff!)
[album]86[/album]
So my next step here is to make a duty cycle controlled fixed frequency PWM stage to see if I can get the full RPM/Power curves…
Stay Tuned!!!
-NerdyEE
P.S. here is a video of it running at 5V output and 1mA into the LED. I measure the RPM but reading needs a /4 due to four holes in the flywheel. I lift the motor and get a typical 1100 RPM…
[youtube]http://www.youtube.com/watch?v=vdlVXZEAwsA[/youtube]
What you proposed is essentially what I have been doing this past week, measuring what my small little JS002 toy SE and the DVD spindle motor can produce in speed and power. I did approximately 20 runs and plotted 20 different RPM/Power curves. Similar trends, but all different data. So your first assumption that the heating source and mechanical are all fixed and do not change, is not a valid assumption…
The concept of the “Dither & Lock” (D&L) uses the response of the SE (bell shaped RPM/Power curve) as an error signal. No need to map out the physical space as the D&L will locate the max point, even if it changes due to pressure, temperature, or phase of the moon… (Even a different mechanical configuration/generator!)
The output of the D&L is a full wave rectified version of the dither frequency (see my Excel simulation). The polarity of this full wave signal will change once past the peak operating point (peak of bell curve). At the peak, the error is zero. Thus taking this error signal into an integrator (PID or the like) can drive the applied loading to the SE to maintain optimum performance. You are correct that this dither frequency needs to be less than the natural response of the SE (need the SE to respond to the change in command for power). From what I see from my toy, that frequency is sub hertz (about 0.1HZ). This may be my difficult challenge as this is a very slow frequency and thus the overall response may be even slower…
FYI- my comment about using ice to kill the toy SE motor, I figured out what was happening. The compression cylinder has an inner sleeve of cast iron (?) and so is the piston. Suspect the people that made this toy read Ian’s numerous posts on friction and the use of cast iron. What happens when I hit the toy with ice is that the compression cylinder seizes solid. But if I cool it slowly (water drips or slow ice ingress) the motor will run very fast. At the time I didn’t have my tachometer so was unable to measure how fast I was able to get the toy. I felt a water drip cooler was not a good long term solution for these tests…
So what have I done lately? I tried to improve the thermal performance of the toy SE. The toy would only allow me about 3 minutes of operation at its peak (> 1000RPM) before it would die off (< 1000RPM). Taking inspiration from Ian’s 2.5cc SE w/ generator, I mounted extra heat sinks and was then able to operate the toy for multiple alcohol bottle burns without a noticeable change in power output. Though honestly, I would rather use Ian’s 2.5cc w/ generator for these experiments as I think my toy is about to die… (have to fix something approx every 5-7 burns and thus obtain different performance)
[album]85[/album]
I think this and my alcohol burner variability is why I can’t obtain good repeatable RPM/Power curves, but I’m going to try it again once I build a true PWM load to measure the full RPM/Power curve (can only measure to the peak now). Somewhere on the forum I read to use graphite as a lubricant. If doing so, I found one has to apply the stuff like eye-shadow (very gently with Q-tip swab and not puff-puff and fly everywhere; a.k.a. MESS!). This does help each re-build and the max “OPEN FLY” performance I have to date is approx 1300RPM (1100 is typical)… (Please note: SE Toy flywheel has 4 holes so tachometer data needs to be corrected by dividing value on meter by 4)
I did find one thing that may be of interest to those who wish to attempt a real Solar MPPT with their SE generator setup. As mentioned before, I ordered a DC-DC converter demo board (Texas Instruments TPS61202EVM-179) that will take the 0.8 – 1.5V output from the SE and DVD Spindle motor combo, and generate a regulated 5V output. I then made a LED constant current source using a LM317 (see picture & videos) to vary the loading on the SE. What I found is that once the power demand from the load was greater than the supply from the SE (approx 12mW), the system would get crow-barred and stall the SE. I somewhat expected this behavior and thus my comments about the use of a Solar MPPT in a previous post… (I can explain what and why this is happening if people are interested in more of my babblings; constant power demand from DC-DC reduces voltage out of SE generator, thus increasing demand for current, thus further reducing voltage, thus further increasing demand for current… Falling off a cliff!)
[album]86[/album]
So my next step here is to make a duty cycle controlled fixed frequency PWM stage to see if I can get the full RPM/Power curves…
Stay Tuned!!!
-NerdyEE
P.S. here is a video of it running at 5V output and 1mA into the LED. I measure the RPM but reading needs a /4 due to four holes in the flywheel. I lift the motor and get a typical 1100 RPM…
[youtube]http://www.youtube.com/watch?v=vdlVXZEAwsA[/youtube]
[hr] If crime doesn't pay, does that mean my job is a crime?
Re: MPPT for Stirling Engines
Hi Vamoose,
I was actually starting to get worried due to my previous experiments that the idea of using a D&L technique to a SE to maintain optimum performance wasn’t going to pan out due to a non-bell shaped RPM –v- Power curve. Well this is now all long gone due to some published performance curves from the Stirling Engine Design Manual, Second Edition by William R. Martini found on Jim Larsen’s amazing web page http://www.stirlingbuilder.com. Buried in there are a bunch of bell shaped curves for RPM –v- Efficiency, RPM –v- Torque; all curve shapes we need to implement a D&L controller. As I read further I found what I needed to see; RPM- v- Power. See screen capture below…
[album]87[/album]
This is good news! Also note the bell shape max changes with pressure, but is still bell shaped. I still need to see if for myself and thus need a fixed frequency (20khz +) PWM stage. I might skimp to save initial time and just use a transistor (MOSFET) and a function generator to see if I can get similar looking curves…
More to come…
-NerdyEE
I was actually starting to get worried due to my previous experiments that the idea of using a D&L technique to a SE to maintain optimum performance wasn’t going to pan out due to a non-bell shaped RPM –v- Power curve. Well this is now all long gone due to some published performance curves from the Stirling Engine Design Manual, Second Edition by William R. Martini found on Jim Larsen’s amazing web page http://www.stirlingbuilder.com. Buried in there are a bunch of bell shaped curves for RPM –v- Efficiency, RPM –v- Torque; all curve shapes we need to implement a D&L controller. As I read further I found what I needed to see; RPM- v- Power. See screen capture below…
[album]87[/album]
This is good news! Also note the bell shape max changes with pressure, but is still bell shaped. I still need to see if for myself and thus need a fixed frequency (20khz +) PWM stage. I might skimp to save initial time and just use a transistor (MOSFET) and a function generator to see if I can get similar looking curves…
More to come…
-NerdyEE
[hr] If crime doesn't pay, does that mean my job is a crime?
Re: MPPT for Stirling Engines
Ok so I’ll make a few clarifications on my previous post. The outline which I described was my initial interpretation of how a Electronic Stirling Engine Load Optimiser might possibly operate, or at least a rudimentary outline of it. But now thanks to J and his dither and lock approach, we could have an advanced concept and method in which to achieve this process. I have confidence his ideas will work, as he really knows his stuff and its based on solid, scientific knowledge and technology (and not the flutterings of an over active imagination).
I also think it would dynamically respond to engines with large Thermal, Pressure and RPM variations as J has indicated, and do this in a more than adequate response time. Which would make it a very handy Stirling engine regulator indeed.
I am still a bit short in fully understanding the specifics of how the system works but think that it creates a kind of moving 'algorithm' or error signal from cross references the rpm response time with the power output, which allows the system to identify and react with a better Load creating another error signal and so on. I could be very wrong about this interpretation of the process. If it sounds like I know what i'm talking about, well I don’t really. Unlike J's 'Dither and Lock', I'm at the edge of my knowledge and am all 'Smoke and Mirrors', I'm much more of a tool in the hand, kind of guy, on some subjects.
Here’s a handy bit of kit (regulator) I've used on renewable energy systems that might have some functions that we are after, and I suspect an Arduino may be able to replicate them in essence.
Its an Australian made PWM regulator (not an mppt though). On the surface it appears quite simple with only one button but when you start to dig it has a lot of handy functions. It has the capacity to actuate multiple relays from different output nodes, for running wind turbine and microhydro generator load dumps and also auto-starts for engines/generators, or in essence to switch anything. It can also run fast switching solid state relays to duplicate its PWM charging process so you can run multiple large arrays off the one small regulator. It has a whole lot of data logging from internal data and from shunt inputs, and many programmable set points and responses. It interfaces easily with a computer through a data cable if one wishes.
Plasmatronics is the name of the company that make it.
They make a few different units but the PL series is the one most commonly used.
I like the 'Dingo' (its basically a PL20 with a few extra inputs/outputs, and is negativly grounded so can be used in cars/vans without confusing people too much)
Here's a link to the Dingo Manual
http://www.plasmatronics.com.au/downloa ... ef.1.3.pdf (about 4MB)
I hope we can get the VMPPT (VamerdyEE (way more NerdyEE than vamoosee) Maximum power point Tracker) running eventually. I hopefully might need one, in time, as I'm about to undertake the building of my first proper engine. The engine design is definitely way out there, so may be a complete flop, but this itch has become one that needs to be scratched.
vamoose
I also think it would dynamically respond to engines with large Thermal, Pressure and RPM variations as J has indicated, and do this in a more than adequate response time. Which would make it a very handy Stirling engine regulator indeed.
I am still a bit short in fully understanding the specifics of how the system works but think that it creates a kind of moving 'algorithm' or error signal from cross references the rpm response time with the power output, which allows the system to identify and react with a better Load creating another error signal and so on. I could be very wrong about this interpretation of the process. If it sounds like I know what i'm talking about, well I don’t really. Unlike J's 'Dither and Lock', I'm at the edge of my knowledge and am all 'Smoke and Mirrors', I'm much more of a tool in the hand, kind of guy, on some subjects.
Here’s a handy bit of kit (regulator) I've used on renewable energy systems that might have some functions that we are after, and I suspect an Arduino may be able to replicate them in essence.
Its an Australian made PWM regulator (not an mppt though). On the surface it appears quite simple with only one button but when you start to dig it has a lot of handy functions. It has the capacity to actuate multiple relays from different output nodes, for running wind turbine and microhydro generator load dumps and also auto-starts for engines/generators, or in essence to switch anything. It can also run fast switching solid state relays to duplicate its PWM charging process so you can run multiple large arrays off the one small regulator. It has a whole lot of data logging from internal data and from shunt inputs, and many programmable set points and responses. It interfaces easily with a computer through a data cable if one wishes.
Plasmatronics is the name of the company that make it.
They make a few different units but the PL series is the one most commonly used.
I like the 'Dingo' (its basically a PL20 with a few extra inputs/outputs, and is negativly grounded so can be used in cars/vans without confusing people too much)
Here's a link to the Dingo Manual
http://www.plasmatronics.com.au/downloa ... ef.1.3.pdf (about 4MB)
I hope we can get the VMPPT (VamerdyEE (way more NerdyEE than vamoosee) Maximum power point Tracker) running eventually. I hopefully might need one, in time, as I'm about to undertake the building of my first proper engine. The engine design is definitely way out there, so may be a complete flop, but this itch has become one that needs to be scratched.
vamoose
Re: MPPT for Stirling Engines
Hi guys, With these very small motors, stay away from electronic convertions, even a diode will drop the voltage .6 of a volt, I know because on my free piston engine with a linear alternator I need to rectify the out put if I want to run a radio.
On the larger motors I find that the power output is fairly flat over quite a wide range of revs, ie., high gearing to the generator, the motor runs slow, but with high torque, orrun with a low gearing, the motor runs at high speed, with low torque. So I tend to find somewere about the middle of the range, and use that. Generally it seems that somewhere about half the free running speed delivers the highest power output. Another example is the stove top fan motor, I can adjust the pitch of the fan, the way I test this is, get the motor running, and in the airstream from it I hang a bit of string, then adjust the fan pitch until the string blows out the most. Not sure on the revs, but I think a bit below half of the maximum the motor could produce if it had a flywheel in place of the fan.
I think that the motor should be designed for its load, or as my ones are, the load is designed for the motor, without electronic intervention. Ian S C
On the larger motors I find that the power output is fairly flat over quite a wide range of revs, ie., high gearing to the generator, the motor runs slow, but with high torque, orrun with a low gearing, the motor runs at high speed, with low torque. So I tend to find somewere about the middle of the range, and use that. Generally it seems that somewhere about half the free running speed delivers the highest power output. Another example is the stove top fan motor, I can adjust the pitch of the fan, the way I test this is, get the motor running, and in the airstream from it I hang a bit of string, then adjust the fan pitch until the string blows out the most. Not sure on the revs, but I think a bit below half of the maximum the motor could produce if it had a flywheel in place of the fan.
I think that the motor should be designed for its load, or as my ones are, the load is designed for the motor, without electronic intervention. Ian S C
Re: MPPT for Stirling Engines
Those pesky little blighter’s, always with their hand out wanting their 0.6. The only way I know to circumvent them is to use a brushed generator and rectify through the commutator, but these generator types have their own inherent problems. I guess you can also transform the voltage up before rectification, but this will induce another set of losses. If someone could devise a way to electronically rectify without having to use a PN junction they would be a very rich person indeed.
I think though Ian, at least in my mind, one of the beneficial uses of this regulator could be to identify the best operational RPM of a Stirling engine electronically so the drive can be ratioed appropriately to know you are getting the the best output from it in whatever mechanical application it may be eventually used for
I do find this observation of much interest Ian
Very interesting observation indeed!
vamoose
I think though Ian, at least in my mind, one of the beneficial uses of this regulator could be to identify the best operational RPM of a Stirling engine electronically so the drive can be ratioed appropriately to know you are getting the the best output from it in whatever mechanical application it may be eventually used for
Although I do not mean to dismiss the potential capacity of the proposed regulator design to be effectively used in the transformation of mechanical Stirling engine power, into usable electrical energy, for larger/higher output engines, if that is their intended use.vamoose wrote:What we're trying to do is find the best load and therefore RPM of an engine in its stabilised long term operating conditions. Whether the 'regulator' overall consumes more power than it makes or vice versa doesn't matter really for the moment. its just a measuring tool. (although it would be nice to get a gain)
I do find this observation of much interest Ian
It has been an assumption of mine that 'up to a certain point', relative to the available rate that heat can be effectively exchanged and also mechanical restrictions, There is a specific rate of processing of the working gas that will give you optimal power. This is to say that I thought every engine, operating under whatever particular conditions at the time, would have a fairly specific optimal RPM for its best power output, and not be spread out over a large window.Ian S C wrote:On the larger motors I find that the power output is fairly flat over quite a wide range of revs, ie., high gearing to the generator, the motor runs slow, but with high torque, orrun with a low gearing, the motor runs at high speed, with low torque.
Very interesting observation indeed!
vamoose
Re: MPPT for Stirling Engines
Gentlemen
An interesting thread 'electronic optimisation of a Stirling cycle engine power curve' and no doubt many problems to be solved. However the main problem to be solved is not control, but 'how to get these engines to produce some useful power in the first place'. After nearly 200 years, a newcomer would quite naturally expect to be able to buy a well sorted engine, off the shelf, and start working on 'electronic improvements', nothing could be further from the truth, there are No reasonably priced engines on the market that are anything other than toys.
There is something very appealing, to me at least and many others it seems, to design a mechanical device, which, when one part of the casing is heated, it produces mechanical work from another part of the casing. Why I continue to spend time trying to crack this problem is beyond me, but I do, and will probably continue to do so until the day they Plant me, but I am under no illision that for such a device to be accepted it has to have more than novelty value.
It may be that for most it is enough just to have a device running on the bench/table top/desk, producing just enough power to overcome internal friction, but not for me. I continue to test ideas in my workshop and have a design in mind which may achieve 'more than novelty value' only time will tell. I just hope that by the time I depart this Planet I am not left thinking, 'Well that was a Bl***y waste of time'.
Keep chugging.
GeoffV
An interesting thread 'electronic optimisation of a Stirling cycle engine power curve' and no doubt many problems to be solved. However the main problem to be solved is not control, but 'how to get these engines to produce some useful power in the first place'. After nearly 200 years, a newcomer would quite naturally expect to be able to buy a well sorted engine, off the shelf, and start working on 'electronic improvements', nothing could be further from the truth, there are No reasonably priced engines on the market that are anything other than toys.
There is something very appealing, to me at least and many others it seems, to design a mechanical device, which, when one part of the casing is heated, it produces mechanical work from another part of the casing. Why I continue to spend time trying to crack this problem is beyond me, but I do, and will probably continue to do so until the day they Plant me, but I am under no illision that for such a device to be accepted it has to have more than novelty value.
It may be that for most it is enough just to have a device running on the bench/table top/desk, producing just enough power to overcome internal friction, but not for me. I continue to test ideas in my workshop and have a design in mind which may achieve 'more than novelty value' only time will tell. I just hope that by the time I depart this Planet I am not left thinking, 'Well that was a Bl***y waste of time'.
Keep chugging.
GeoffV
Re: MPPT for Stirling Engines
Geoff, these things are addictive, although proberbly safer than fags. I find that even the "toy" Stirling Engines are quite expensive here, and that poor wee thing would not be any use trying to power something, although it might turn a generator, and light a LED.
I like something I can hang a brake on, get the inch onces, the rpm, then load up the brake, and do it again. Then hook up the generator with a load, a volt, and amp meter, and a hand full of different sized pullies, its all amps/milli amps X volts= watts. Ian S C
I like something I can hang a brake on, get the inch onces, the rpm, then load up the brake, and do it again. Then hook up the generator with a load, a volt, and amp meter, and a hand full of different sized pullies, its all amps/milli amps X volts= watts. Ian S C
Re: MPPT for Stirling Engines
Hey Folks,
Thanks for your interest and input.
These engines definitely get under ones mechanical skin.
I think that all our yolks might be slightly offset, and shafts bent, to varying degrees.
Ian, I hope we don’t have to take up Durries (smokes) to cure ourselves of this affliction.
Although a newbie to Stirling engines, I'm under no illusions of the general limited availability of engines with any respectable power, out there in the world, say for a few exceptions, of the same ilk that u and a few others are researching Geoff, and the 'Expensive' ones (but maybe you, or they, could have a use for such an electronic regulating device).
It would be an interesting exercise to run a comparison between the proposed regulator we are discussing (if it comes into being) and the traditional hands on mechanical method of analysis, to see if the device is able to achieve what we are attempting to do.
Also, for smaller engines, it would be a nice little doodad to have in ones testing kit, for cross analysis of the engine 'for interests sake' (if it works)...
For myself the regulator concept, is in part an exercise, kind of for the sake of it (what’s over that hill..). Its fairly out of my depth, but I feel its J's forte, and I think he also wants to use the thread for a possible electronics type teaching tool, from what I’ve read, (although I don’t mean to speak for him). I hope he wishes to continue with it, if he finds the time. I am definitely a keen observer and hope to learn a lot in the process, as the idea progresses.
I am accumulating the materials and am just about to buy a tig welder, to attempt the making of my first engine (it may take some time). Its my own design and have not seen anything similar in my researching (maybe for good reason). It will be a good test to see if I really have my head around the hot air engine concept and practicalities, in a proper sense. Fingers and pistons crossed...
vamoose
Thanks for your interest and input.
These engines definitely get under ones mechanical skin.
I think that all our yolks might be slightly offset, and shafts bent, to varying degrees.
Ian, I hope we don’t have to take up Durries (smokes) to cure ourselves of this affliction.
Although a newbie to Stirling engines, I'm under no illusions of the general limited availability of engines with any respectable power, out there in the world, say for a few exceptions, of the same ilk that u and a few others are researching Geoff, and the 'Expensive' ones (but maybe you, or they, could have a use for such an electronic regulating device).
It would be an interesting exercise to run a comparison between the proposed regulator we are discussing (if it comes into being) and the traditional hands on mechanical method of analysis, to see if the device is able to achieve what we are attempting to do.
Also, for smaller engines, it would be a nice little doodad to have in ones testing kit, for cross analysis of the engine 'for interests sake' (if it works)...
For myself the regulator concept, is in part an exercise, kind of for the sake of it (what’s over that hill..). Its fairly out of my depth, but I feel its J's forte, and I think he also wants to use the thread for a possible electronics type teaching tool, from what I’ve read, (although I don’t mean to speak for him). I hope he wishes to continue with it, if he finds the time. I am definitely a keen observer and hope to learn a lot in the process, as the idea progresses.
I am accumulating the materials and am just about to buy a tig welder, to attempt the making of my first engine (it may take some time). Its my own design and have not seen anything similar in my researching (maybe for good reason). It will be a good test to see if I really have my head around the hot air engine concept and practicalities, in a proper sense. Fingers and pistons crossed...
vamoose
Re: MPPT for Stirling Engines
Vamoose
Sorry to hear that you have also contracted 'The Bug' but you're in good company, so welcome to the Cul de Sac club.
Your endevours with an electronic control system may well be of value in the future, my concern is, at the moment, it's rather like trying to optimise the thrust generated during a bout of flatulence!
Good to hear you're about to start cutting metal, but if I could offer some general advise before you enter your first Cul de Sac.
1. Don't beleive a word from the Academics until they have demonstrated their theories.
2. Test parts of your design on the bench before proceeding, the lack of heat transfer to the working gas is much less than expected.
3. If your design has the bottom of the piston/s open to a crankcase, beware of large 'Forced Work' losses.
4. Don't expect to get any usefull power without pressurisation, unless the engine is huge.
Good luck.
GeoffV
Sorry to hear that you have also contracted 'The Bug' but you're in good company, so welcome to the Cul de Sac club.
Your endevours with an electronic control system may well be of value in the future, my concern is, at the moment, it's rather like trying to optimise the thrust generated during a bout of flatulence!
Good to hear you're about to start cutting metal, but if I could offer some general advise before you enter your first Cul de Sac.
1. Don't beleive a word from the Academics until they have demonstrated their theories.
2. Test parts of your design on the bench before proceeding, the lack of heat transfer to the working gas is much less than expected.
3. If your design has the bottom of the piston/s open to a crankcase, beware of large 'Forced Work' losses.
4. Don't expect to get any usefull power without pressurisation, unless the engine is huge.
Good luck.
GeoffV
Re: MPPT for Stirling Engines
I'm not trying to start a flame war here, but I have issues with this statement.Geoff V wrote: snip...
1. Don't beleive a word from the Academics until they have demonstrated their theories.
snip...
GeoffV
Firstly using the word academics with capitalization has an implication which places scientists/academics in a category similar to Catholics or Mormons (a sort of religious effort). Learning, and teaching, should be respected efforts and such negative connotations indicates a prejudice against such activity whether intended or not. Without these academics we would still be bleeding folks as a valid medical treatment, and the net wouldn't exist.
Secondly the word theory is used, here, in a manner inconsistent with the proper definition. Academics, or any interested party, may test a hypothesis, or demonstrate the validity thereof, whereas a theory has already survived such testing and demonstration. Theories may still be overturned, or revised, but only by testing which provides consistent alternative results. Thermodynamics is a very deeply examined matter and has withstood all sorts of testing to the point that the theory of thermodynamics is accepted as fact.
Finally the acceptance of the results of testing a hypothesis to the point that a theory emerges isn't a matter of faith. One can either accept testing results or reject them, but rejection out of hand is more an act of belief than trusting reports submitted by academics. Why? Academics publish their findings in a manner which allows anyone to review these findings, and test the results for themselves. Belief plays no part in true academic endeavors, and this holds true for all of science.
Having been employed as a professional scientist, and being exposed to academic practices, I feel it important that these issues be addressed. I'm NOT claiming that all such efforts are perfect. Far from it. What I'm saying is that academics, and scientists, provide us a rigorous means by which we may examine and test their efforts for ourselves. Nearly every error in such efforts have been exposed by other academics and scientists, and corrected.
I get very defensive when inaccurate descriptions of these efforts are put forward and I see it as my duty to correct such inaccuracies. It may have been you intention to say pretty much the same thing as I'm saying here, but I feel it is vital that folks unfamiliar with such issues be exposed to the exacting practices under which academia and science operates. Proper terminology, and our understanding of such terms, is vital for useful communication.
R
Re: MPPT for Stirling Engines
theropod2
I appear to have hit a nerve, good, I've been mislead for 30 years by authors publishing theories as to how the Stirling cycle functions, it wasn't until I disregarded these learned words and started to bench test every possible aspect of the design that I started to make some progress with these engines. It was the fine engineer Kieth Duckworth of Cosworth fame who wrote in his book 'it is better to be uninformed than misinformed' a sentiment I totally agree with. So I'm sorry if I have offended you, but if I can save other engineers from wasting time and money through following the untested theories of those who think they know better, I will.
Vamoose
One thing I forgot to mention, computer simulations from my experience are also a complete fantasy even when the data and test results of an actual engine are entered. In the theoretical world and the simulation world, cylinders are perfectly round and parallel and remain so through the operating temperature range of the engine. In the real world they are neither round nor parallel except, occasionally, at one temperature, hence my offered advise to test each aspect of your design to measure what it is actually doing.
Have a nice day.
GeoffV
I appear to have hit a nerve, good, I've been mislead for 30 years by authors publishing theories as to how the Stirling cycle functions, it wasn't until I disregarded these learned words and started to bench test every possible aspect of the design that I started to make some progress with these engines. It was the fine engineer Kieth Duckworth of Cosworth fame who wrote in his book 'it is better to be uninformed than misinformed' a sentiment I totally agree with. So I'm sorry if I have offended you, but if I can save other engineers from wasting time and money through following the untested theories of those who think they know better, I will.
Vamoose
One thing I forgot to mention, computer simulations from my experience are also a complete fantasy even when the data and test results of an actual engine are entered. In the theoretical world and the simulation world, cylinders are perfectly round and parallel and remain so through the operating temperature range of the engine. In the real world they are neither round nor parallel except, occasionally, at one temperature, hence my offered advise to test each aspect of your design to measure what it is actually doing.
Have a nice day.
GeoffV